VoiceVector is a voice technology platform that offers three types of services: voice cloning, text-to-speech (speech synthesis), and speech-to-text (speech recognition). Its official website states that these technologies are free to use (cutting-edge technology) and adopt a flexible pay-as-you-go model. The platform says it is suitable for developers, podcasters, and content creators.
Uses The platform is aimed at users who need to generate or convert voice content, and can be used to generate audio for text, convert speech to text, and clone a personal or specified voice for text-to-speech.
Main capabilities
- Voice cloning: Submit 1-2 minutes of audio to train a model. The official website states that a cloned voice can be obtained in about 10 minutes. Any number of voices can be cloned and retained, with consistent voice reproduction.
- Text-to-speech: Convert any text into natural audio, providing more than 100 built-in voices covering 20 languages. You can also use your own cloned voice, and long text is supported.
- Speech-to-text: Convert spoken language into text, suitable for meetings, lectures, and everyday communication. It supports more than 100 languages and variants, provides high-quality transcription, and allows downloading srt or txt files.
How to use After registering, users can get $2 in free credit and 3 free voice clones. Voice cloning requires only 1-2 minutes of audio; text-to-speech is billed by character and allows downloading wav or mp3 files; speech-to-text is billed by minute and allows downloading srt or txt files, and includes 400 minutes of free transcription time and 7,692 free characters for speech synthesis. When using your own trained model for text-to-speech, the charge is $0.26 per 1,000 characters.
Payment models
- Pay-as-you-go: Pay according to actual usage, with a minimum top-up of $5 and no limit on the number of top-ups or the maximum amount. Text-to-speech is billed by character, rounded to the nearest $0.001; speech-to-text is billed by minute. Credit does not expire, but is non-refundable.
- Subscription: Suitable for users with higher usage, allowing up to 200 hours of audio transcription, 20 voice clones, and 120,000 characters of text-to-speech per month. Subscription and pay-as-you-go can be used at the same time. Subscription credit is deducted first, and pay-as-you-go credit is used after the subscription credit is exhausted. If pay-as-you-go credit remains, it can be used to offset the subscription fee, with a minimum subscription payment of $1 (for example, for a $20 subscription, if there is $19.5 in pay-as-you-go credit, $19 is used to offset it, and only $1 needs to be paid).
Applicable scenarios and limitations
- Applicable scenarios: generating audio for articles, product descriptions, educational content, etc.; meeting and lecture transcription; podcasting and content creation;developer integration of voice capabilities.
- Limitations: Cloned voices are private to the account only and cannot be shared with other accounts; there is no limit on the total number of cloned voices retained in an account, but subscription plans limit the number of new clones per month; cloned voices that have not been used for text-to-speech for more than 2 months will be deleted; after canceling a subscription, cloned voices will still be retained as long as they are used for text-to-speech at least once every 2 months; cloned voices can be used for commercial purposes, but you must ensure that you have the legal right to use that voice for commercial purposes, and the platform does not assume legal liability; failed requests are not charged; the service quality of subscription and pay-as-you-go is the same.
The above information all comes from official public materials. For specific features, prices, and availability, please refer to the latest instructions on the official website.
